The World’s Largest Open Access Agricultural & Applied Economics Digital Library

This document is discoverable and free to researchers across the

globe due to the work of AgEcon Search.

Help ensure our sustainability.

Give to AgEcon Search

AgEcon Search
http://ageconsearch.umn.edu
aesearch@umn.edu

Papers downloaded from AgEcon Search may be used for non-commercial purposes and personal study only.
No other use, including posting to another Internet site, is permitted without permission from the copyright
owner (not AgEcon Search), or as allowed under the provisions of Fair Use, U.S. Copyright Act, Title 17 U.S.C.
 Economic Evaluation of Achieving Biofuel Mandate through Advanced Biofuels in Developing
Country: Case of India

by

Prantika Das1 and Haripriya Gundimeda2

Copyright 2021 by Prantika Das and Haripriya Gundimeda. All rights

reserved. Readers may make verbatim copies of this document for non-
commercial purposes by any means, provided that this copyright notice
appears on all such copies.

1
Department of Humanities and Social Sciences, Indian Institute of Technology, Bombay, India
2
Professor, Department of Humanities and Social Sciences, Indian Institute of Technology, Bombay, India
 Economic Evaluation of Achieving Biofuel Mandate through Advanced Biofuels in
Developing Country: Case of India

Abstract

India’s biofuel blending has fallen short of goals since its implementation in 2003. Policy
regulation for using only non-food and non-arable land for feedstock production have caused
inadequacy in feedstock availability raising mandate compliance challenges. India’s New
Biofuel Policy, 2018 allowed expansion of feedstocks to include agricultural residues for
biofuel production. As a first attempt, this study aims to assess biofuel impact with cellulosic
biomass in a unified framework of food and fuel for India. The study intends to identify the
optimal feedstock mix and assess the potential of cellulosic biofuels in achieving 20% blending
mandate by 2030. We use price endogenous mathematical programming model to analyse
impacts of alternate feedstock combinations on food, land use and economic surpluses of
food and fuel consumers and producers. The study undertakes analysis of three policy
scenarios that allows alterations induced by different feedstock combinations. The findings of
our study suggest that with limited scope for area expansion, cellulosic ethanol can potentially
help dampen the pressure of land expansion for sugarcane crop as compared to situations
where the mandate is met with only sugarcane-based feedstocks. Our results indicate that
cellulosic biomass producers can enjoy significant economic benefits from sale of agriculture
residues.
Key words: Biofuel policy, biomass, Partial Equilibrium Model

JEL Codes: Q160, Q180, C610

Acknowledgement: This work has been financially supported by the Department of Biotechnology,
India (Grant: DBT/EB/PANIIT/2012).
1. Introduction

India’s endeavour towards biofuel is guided by multiple policy goals. With a relatively high
crude oil import dependency of about 84% in 2018-19 (PPAC, 2019), India aims to achieve
energy security by reducing import dependencies by 10% by 2022 through biofuels as one of
the identified approaches (MoPNG, 2018). In addition to achieving energy security through
fossil fuel replacement, biofuels are also of strategic importance in India’s endeavour to
double farmers’ income, employment generation and GHG reduction targets. India has been
aggressively promoting biofuels with the implementation of National Policy on Biofuels, 2009
through an indicative blending mandate of 20% for both ethanol and biodiesel targeted to be
achieved by 2017. Nevertheless, India’s biofuel blending obligations has always fallen short of
goals since its implementation.

Risks of biofuel induced food insecurity and land use change have seriously hindered growth
of biofuels in developing countries (Msangi and Evans, 2013). However, Indian policy makers
have pursued biofuels with caution (Findalter and Kandilkar, 2011; Ray et al., 2012). To avoid
any negative market mediated impacts, India’s biofuel policy relied only on use of non-edible
feedstocks and non-arable land for feedstock cultivation i.e., molasses ethanol (a by-product
of sugarcane processing) and Jatropha cultivated on marginal lands (MoPNG, 2009). Despite
of such unique approach in biofuel promotion, multiple factors like feedstock supply shortage,
competing demand for ethanol for other industrial uses, inefficient pricing mechanism of
ethanol, restricted feedstock options, regional administrative challenges in implementing the
mandate and mismatch between market conditions for ethanol, sugar, and petrol lead to the
failure of Ethanol Blending Programme (EBP) (Saravanan et al., 2018; Ray et al., 2011). Even
biodiesel production from Jatropha never really took off commercially due to inefficient
supply-chain, lower Jatropha yields, and miscalculation of potential availability of wasteland
(Kumar et al., 2012; Axelsson et al., 2012; Montobbio and Lele, 2010), overtly leading to the
failure of National Biodiesel Mission (NBM). Biofuels market penetration rate in 2019 was
only about 5.8% for ethanol with consumption of about 2400 million litres and 0.14% for
biodiesel with a consumption of about 85 million litres (GAIN, 2019).
India allowed expansion of feedstock options to include cellulosic feedstocks from agricultural
residues for ethanol production with the formulation of the New Biofuel Policy, 2018. The
new policy promotes advanced biofuels from agricultural biomass, direct sugarcane juice
from excess sugarcane produces and other feedstocks which are in surplus availability.
Ethanol from agricultural residues can prove to be a solution to India’s longstanding residue
burning problem. In addition, ethanol from sugarcane and bagasse can potentially help in
revenue diversification for its financially strained sugar industry (Ray et al., 2012; Lavanya and
Manjunatha, 2018). With around 611 MT to 686 MT of annual gross residue production of
which nearly 25%-34% is available as surplus for bioenergy purposes (Hiloidhari et al., 2014;
Cardoen et al., 2015), if implemented sustainably and efficiently, biofuels from agricultural
residues can prove to be a win-win condition for both farmers and biofuel producers of India.
There is a growing body of literature on biofuel impact assessment on food prices, land use
change and GHG emissions. However, most of the studies have visibly remained biased
towards developed country, with an emphasis on U.S., EU biofuel policies. Literature review
indicates a dearth of impact assessment studies for India. Existing studies are limited to first
generation feedstocks for ethanol and non-edible oilseeds for biodiesel (Khanna et al., 2013;
Gunatilake et al., 2014; Woltjer and Smeets, 2016). These studies indicate inefficacy of
molasses feedstock and strong ability of Jatropha on marginal lands in meeting the national
blending targets. However, to the best of our knowledge, none of the study has attempted to
evaluate the differential impact of ethanol from cellulosic biomass on food and land use for
India. The primary objective of the study is to identify the optimal feedstock mix in achieving
20% ethanol blending mandate by 2030 and analyse its impact on food and land use, prices,
and the welfare impacts on food and fuel consumers and producers. The study also aims to
assess the potential of cellulosic biofuels in achieving India’s biofuel obligation. Biofuel policy
scenarios in the study helps assess alterations induced by changes in feedstock mix to attain
the 20% blend requirement by 2030. The focus here is on evaluating the feasibility of
achieving the biofuel mandate with alternate mix of first and second-generation feedstocks.

2. Overview of biofuel policy in India

India initiated its effort to develop a sustainable biofuel industry since the early 2000s. The
country introduced a 5% ethanol blending pilot program in 2001 which was later transformed
into 5% blending mandate applicable to nine states and four Union Territories (UTs) in 2003,
which was then expanded to the whole country except for few north-eastern states in 2006.
This was accompanied by National biodiesel Mission (NBM) in 2003 with a 20% biodiesel
blending target, set to be achieved in a phased manner by 2011-12. However, faced with
serious supply shortages of feedstocks for both ethanol and biodiesel, India’s biofuel policy
never really took off the planned way. Finally, the National Policy on Biofuels (NPB) was
implemented in 2009 (hereafter referred as NPB, 2009) with a proposal of achieving 20%
blending mandate for both ethanol and biodiesel by 2017 (MNRE, 2009). The NPB, 2009 was
implemented with a unique concept of biofuel promotion with indigenous non-food
feedstocks grown on non-arable land with two integral components of the policy i.e., Ethanol
Blending Program (EBP) and the National Biodiesel Mission (NBM). Intending to avoid any
food-security concerns and to fully utilize the already established ethanol production capacity
by distillers for potable liquor and industrial use, the EBP was initiated with molasses
feedstock which is a by-product of sugar processing. The NBM pursues biodiesel3 processing
from Jatropha oilseeds, supposed to be grown on non-arable land or wasteland.
With continuous government interventions, ethanol blending in India caught some pace since
2014. Due to favourable pricing conditions, the all-India ethanol blending rate increased from

3
Since in this paper we have only included ethanol, policy overview for biodiesel is not discussed in detail here.
1.8% in 2011 to 3.2% in 2014-15. Figure 1 presents the fuel ethanol use and blend rate
achieved over the years.
Figure 1: Fuel Ethanol Use and Blending Rate India, 2009-2018
Fuel Ethanol Use and Blend Rate Achieved
1400 3.3 3.2 3.5
1200 3
1000 2.3 2.5
1.8 1.9
800 1.6 2
1.4 1.4
600 1.5
400 0.6 1
200 0.3 0.5
0 0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Fuel Ethanol (million litres) Blend Rate (%)

Source: Biofuels Annual, 2018, GAIN

The two major policy issues in India’s biofuel sector are availability of alternate feedstock
options, and pricing mechanism of biofuels. Therefore, amending the previous policy, the
renewed National Policy on Biofuels was initiated in May, 2018 (hereafter referred as NPB,

2018) aiming to achieve an indicative blending target of 20% ethanol and 5% for biodiesel by
2030. Effective realization of the policy is based on five identified approaches i.e., enhancing
domestic supplies, setting up 2G biorefineries, new feedstock development, developing new
technology for biofuel conversions, and creating enabling environment for its integration with
conventional fuel. Efforts to exploit all possible first-generation feedstock option that
including B-heavy molasses, direct sugarcane juice, and damaged food grains like wheat,
broken rice etc. which are unfit for human consumption and surplus food grains in years of
excess production will also be made through the new policy,
3. Objective of the study

The primary objective of the study is to assess policy scenarios in coherence with the National
Policy on Biofuels 2018 focussing only on ethanol. Few critical policy questions that arise
related to India’s ethanol blend mandate are- 1) Is it feasible to meet 20% ethanol blending
mandate by 2030? In accordance, what are the optimal feedstock mix required to meet the
blending mandate for 2030 and what will be the effect of different feedstock mix on food and
land use? 2) How can cellulosic ethanol potentially contribute towards achieving the blending
mandate by 2030? 3) What are the distributional impacts of alternate feedstock mix on
different stakeholders i.e., producers and consumers of food and fuel, and the government.
This study aims to answer these critical policy questions using a price endogenous
mathematical programming model.

4. Methodology
The method used for evaluation is theoretically analogous to several studies undertaken for
agricultural policy analysis under market equilibrium condition. Specifically, the study uses
price endogenous mathematical model as an analytical tool to evaluate the above research
questions. The conceptual underpinning of the model is based on the seminal work of
Samuelson (1952) which was later enhanced and reformulated by Takayama and Judge (1971,
1964 a, c) into a quadratic programming formulation. McCarl and Spreen (1980) extensively
discusses the microeconomic foundations and underlying premises of quadratic programming
models wherein each producing unit individually behave as small competitive units, but
collectively, prices and quantities are endogenous to the model. The modelling approach has
been widely applied for environmental and agricultural policy analysis and bioenergy impact
assessment. Specific to India, Khanna et al., (2013) used price endogenous partial equilibrium
model for assessing impact of meeting ethanol mandates with first generation feedstocks.
This study builds on the modelling framework of Khanna et al., (2013) and uses Chen et al.,
(2014) and Nunez et al., (2013) frameworks to allow for inclusion of cellulosic ethanol and
ethanol from other first-generation feedstocks. For this purpose, we develop a
multicommodity, price endogenous, partial equilibrium model with an objective function
maximizing society’s welfare. The quasi-welfare function is represented by the sum of
consumers’ and producers’ surplus derived from production and consumption of several
agricultural commodities, fuel, and biofuels. The model includes spatially explicit agricultural
supply functions and defines commodity demand at all India level. The model distinguishes
between biofuel types produced from three different feedstocks i.e., molasses, sugarcane
juice and cellulosic biomass. Specifically, the study includes three different sources of biomass
i.e., rice straw, wheat straw and bagasse.
The supply response of agricultural sector is represented by a Leontief production function in
which land as the only primary input, while availability of all other inputs (e.g., fertilizers,
pesticides, irrigation, labour etc.) is assumed to be unlimited at constant prices observed in
the base year. Land use in each region is restricted to the total cropland available in the base
year. Historically observed allocation of land in each region among the crops are used to
constrain land use decisions to avoid abrupt and unrealistic changes in land allocation
(McCarl, 1982; Onal and McCarl, 1989). The model operates under perfectly competitive
market, in which each individual producer are price takers. The consumer demand functions
for agricultural commodities are explicitly defined by price-quantity relationship using the
base year price and quantity level and elasticity of demand which is assumed to remain
constant for future period. The domestic and export demand function for commodities and
export supply functions are assumed to be linear and separable. Demand for ethanol is
expected to shift the demand curve for sugarcane and other biofuel feedstock crops (rice and
wheat for straw) to the right.
4.1. Data description and assumptions
The data for the model comes from 16 major agricultural supply regions, together
representing nearly 96% of India’s gross cropped area and nine different crops. The selected
nine crops include rice, wheat, sorghum, maize, groundnut, rapeseed, soybean, and cotton.
The model currently considers only sugar4 as the processed commodity. Regional
heterogeneity in the model is explained by state specific data on available land resources,
crop productivity, costs of crop production and biomass yield. Data inputs for model
simulation includes base year data for domestic demand and export-import quantity and
prices of modelled commodities, historical crop production possibilities of selected crops,
crop yields, crop production costs, technological costs for crop processing and biofuels
processing, biomass yields, residue-to-crop (RCR) ratios, and processing factors of different
processed commodities including biofuels.
Costs of production of row crops include operational costs and fixed costs. Operating costs
are yield dependent and include costs of labour (human labour, animal labour and machinery
costs), seeds, fertilizer and manure costs, irrigation charges, insecticides, interest on working
capital and other miscellaneous costs. The study uses three-year average (2009-2011 adjusted
to 2011 prices) costs for each cost component to represent crop production costs at the base
year. Data for crop production costs is obtained from crop production budgets compiled for
each state through the Cost of Cultivation Surveys published by Directorate of Economics and
Statistics (DES), Ministry of Agriculture (MoA). In the model, land allocation decisions are
endogenous and changes in cropland allocation is determined by relative prices of the crops.
Cropland availability limits the total regional cropland usage to the available hectares of land
in that region i.e., limited to the initial (base year) agricultural endowment of cropland. State
specific planted areas for the nine selected crops for period 2000 to 2011 was obtained from
DES. The data obtained was used to construct the cropland available in 2011.

We use Khanna et al., (2013) crop calendar approach to specify cropping activity pattern for
different regions and crops. Crop calendar information for each state and crop was obtained
from the Indian Council of Agricultural Research (ICAR) and MoA. Crop-mixes have been
restricted to 2001-2011 period. The model includes crop-mix restrictions on a state basis
covering all the major crops in the model. Historical data for crop yield is obtained from DES.
The model uses state-wise historical three-year average (2009-2011) yield per hectare for
crops by agricultural seasons as the representative yield for that state for specific cropping
seasons. Domestic demand quantity, prices and elasticity parameters are the key inputs
calibrating the demand curves for commodities in base-year. The base year demand
quantities were obtained from USDA and CACP and represent an average of 2009-2011
quantities. Domestic demand quantities represent total demand and includes food and feed

4
In India, though sugarcane is primarily used for sugar production, there are also alternate uses like manufacturing
of Gur, Khandsari, seed and others. The model does not capture the market of Gur and Khandsari.
demand. Domestic prices for agricultural commodities are represented by farm harvest
prices. The model uses average of 2009-11 farm harvest prices (adjusted to a constant 2011
prices). Price elasticities of commodities were obtained from different sources.
The model includes two major ethanol types i.e., ethanol made from sugar-based feedstocks
and cellulosic ethanol. The model allows hydrous ethanol production only from molasses
while anhydrous ethanol can be produced from all feedstocks included in the model. The
model considers regional heterogeneity in technical parameters like sugarcane processing5,
molasses recovery rates and sugar recovery rates. Cellulosic feedstock includes rice and
wheat straw and sugarcane bagasse. Yields and costs of production of crop residues are region
specific and which were computed based on the RCR ratios, moisture content and residue
collection rates which are dependent on alternate residue uses. In India, conservation
agriculture (CA) adoption is still in the initial phases (Bhan and Behra, 2014) and rice-wheat
cropping system in India involves intensive tillage along with complete removal of crop
residues. Therefore, the model assumes that entire land under respective feedstock crop is
available for residue collection and accordingly no residue is used for in-situ uses. However,
the model accounts for competing exsitu uses of residues and uses residue utilization rates
reported in Cardoen et al. (2015b). Cost of agriculture residues in the model includes costs of
residue production, harvesting, collection, transportation and storage of agricultural residues,
all evaluated using Tripathi et al. (1998) approach. We assume manual collection of residues.

In the fuel module, we assume a linear and downward sloping inverse demand function for
fuel (blended gasoline and ethanol). The aggregate demand for fuel is defined to include both
anhydrous ethanol and gasoline (petrol) which was around 21519 million litres in 2011 of
which only 365 million litres were ethanol. Retail prices for gasoline was obtained from Indian
PNG Statistics (2011). For calibrating the demand curve of blended fuel, the study uses Gorter
et al., (2015) and Khanna et al., (2016) approach. Ideally, the consumer price of blended fuel
is the weighted average of anhydrous ethanol and gasoline prices (inclusive of taxes), where
the weights are the shares of ethanol and gasoline in the final fuel blend. We use the central
and state tax information on gasoline and ethanol along with marketing margins and other
taxes to get the retail prices of blended fuel. The tax rates assumed here are based on rates
levied in the state of Delhi in 2011. Finally, the demand curve for blended fuel is defined by
using the weighted average of gasoline and ethanol prices and quantity and a fuel demand
elasticity. Since, blended fuel is not segregated with non-blended fuel at pump, we assume
elasticity of gasoline to be same as elasticity of blended fuel. Specifically, the study uses
gasoline price elasticity of -0.36 as reported in Dahl, (2012). Since, India’s vehicle fleet
structure is constrained from operating on high ethanol blends, we assume that gasoline and
biofuels are imperfect substitutes in producing the blended fuel. Since trading of ethanol for

5
In India, sugarcane harvested is not equal to sugarcane crushed for sugar purpose. Since, the study does not
include gur and Khandsari in the processed commodity list, sugarcane use for processing has also been adjusted
accordingly (sugarcane use for gur and khandsari excluded). The proportion of sugarcane crushed for sugar varies
by region depending upon the importance of sugar-based products (sugar, gur, khandsari, seeds etc.) in specific
region.
fuel blending is prohibited by policy in India, the model does not allow trading of ethanol for
fuel blending, however ethanol trading for industrial uses is allowed in the model.

The model is calibrated using 2011 as the base year. For simulating the 2030 case, we use
exogenous shocks to update the model. The model projected forward to 2030 i.e. the year by
which NPB, 2018 policy target is expected to be achieved. The crop yields are assumed to
increase over time at an exogenously specified rate estimated using historical data for the
period 1980-2014. A critical assumption made over here is that although each region (states)
may have different yield in the base year, they are all expected to grow at the same annual
growth rate. In the fuel sector, we assume that the historical trend of petrol demand i.e.,
average annual growth rate of 8.5% between 2001-2018 (PPAC) to continue between 2011-
2030. Domestic demand for commodities is assumed to change over time based on historical
growth trend and future requirements. To simulate for 2030, domestic demand curves are
shifted upward over time at exogenously specified rates. The shift parameters or the annual
growth rates were estimated based on OECD-FAO Agricultural Outlook projections from 2014-
2023 which provides country wise growth projections in agricultural exports, imports,
production and consumption. For alcohol, due to unavailability of any recent data on future
demand for industrial ethanol (alcohol) we use the alcohol demand growth rate estimates of
other studies. According to Shinoj et al., (2011) and Purohit and Fischer (2014) the industrial
ethanol demand growth rate increased by 3% and potable ethanol by 3.3%, and the trend is
likely to remain same in the future. Since, the model considers combined demand for alcohol
(industrial ethanol and potable ethanol), the projected demand is expected to grow at an
annual rate of 6.3%. For the 2030 case, expansion of agriculture production is modelled as a
product of yield improvements on the agricultural land currently in use. The equilibrium price
of ethanol in the national market, the amount of land cultivated and the prices and quantities
of sugar, alcohol, and all primary agricultural commodities are determined endogenously by
the model.
The mathematical representation of the model can be found in the Appendix. The model is
coded using the Generalized Algebraic Modeling System (GAMS) language.
5. Results
5.1. Description of biofuel scenarios
Policy scenarios for assessment have been designed in coherence with the National Policy on
Biofuels 2018, which allows alterations induced by changes in feedstock mix to attain the 20%
ethanol blending requirement by 2030. Specifically, we assess three policy scenarios engaging
alternate feedstock mix. The first scenario considers 20% ethanol blending mandate by 2030
to be achieved through only molasses feedstock. With insufficient advancement in availability
of alternate feedstocks, it is expected that ethanol would continue to be produced primarily
from molasses. This scenario is referred as the ‘Baseline Scenario’ from here on. We compare
this 2030 baseline with two alternative policy scenarios. The new biofuel policy allows use of
surplus food grains for production of ethanol including sugarcane. With this policy sanction,
despite of the fact that producing ethanol from sugarcane will require major capacity
expansion it is quite evident that in the surplus production years sugar mills will divert
sugarcane towards ethanol production. While ethanol supply from molasses is likely to
continue, this scenario referred as “Combined Molasses & Sugarcane” scenario considers the
case in which the 20% blending requirement is to be met through a mix of molasses and
sugarcane feedstock. To avoid unrestricted diversion of sugarcane for ethanol production,
this scenario considers the case in which by assumption only the leading sugarcane producing
states (i.e., U.P., Maharashtra, Karnataka, and Tamil Nadu) can divert 10% of the sugarcane
for ethanol production. An additional scenario for unrestricted sugarcane uses for ethanol has
also been tested. The second alternative scenario referred as “All-feedstock” scenario
considers the case in which the 20% ethanol blend mandate is supposed to be met with
cellulosic feedstocks in combination with molasses and sugarcane juice. The scenario
particularly considers ethanol production from bagasse, rice straw and wheat straw. mandate
with

5.2. Model validation

As a first step in testing for model performance, the model is first validated for year 2011 and
a comparison is drawn for the simulated model on land allocation, crop production, fuel use
and commodity prices with the corresponding observed values in the base year i.e., 2011.
Domestic transportation of crops to consumers is not considered for model simulation and
therefore, the crop prices solved by the model represent producer prices at the farm-gate.
Further, international transportation costs are also not considered in case of exports and
imports and therefore, it is quite likely that the model is mis specified. To overcome
misspecification error, we follow Fajardo et al., (1981) approach, in which difference between
domestic and foreign prices of commodities are considered as marketing margins which is
then adjusted in the objective function.
Table 1 summarizes the validation results for land use and fuel sector. The land use of
agricultural crops is within acceptable margin of errors. The fuel consumption values are also
within a small deviation range around the observed values. This indicates good simulation
performance of the model. These small deviations notwithstanding, the model appears to
reasonably replicate the base year market equilibrium conditions in the food and fuel sectors.
Table 1: Model Validation results for land use and fuel sector
Observed Base Year Model Base Year Deviation (%)

Land Use (in million Ha)

Rice 40.9 43.0 5

Sorghum 6.2 6.4 2

Maize 7.8 7.6 -1

Groundnut 5.2 5.5 5

Wheat 29.0 28.7 -1

Soybean 10.1 9.9 -1

Rapeseed 5.5 6.7 22

Cotton 12.1 12.0 -1

Sugarcane 5.0 4.9 -2

Fuel Sector

Gasoline Prices 65 64 -1.2

(Rs./litre)
Gasohol Prices 65 64 -1.7
(Rs./litre)
Alcohol Prices 43 43 0.0
(Rs./litre)
Gasoline 21154 21298 0.7
Consumption (million
litres)
Ethanol Use (million 365 372.7 2.1
litres)

5.3. Feasibility of achieving 20% blend mandate by 2030- identifying the optimal feedstock
mix

5.3.1 Only molasses ethanol 2030: In this scenario, the results suggest that only 2810 million
litres of ethanol can be produced from molasses in 2030. The maximum achievable blending
rate with only molasses feedstock is only 2.92%. Molasses production of about 12.8 million
tonnes is likely to be insufficient to meet the increasing demand for both fuel blending and
industrial alcohol production. Due to increased demand of molasses for fuel purpose all
molasses produced is diverted to fuel ethanol production as a result of which domestic supply
of alcohol falls to almost zero and therefore entire domestic requirement of alcohol is met
through imports (5475.152 million litres). No significant increase in area under sugarcane is
observed as scope of area expansion under sugarcane is limited due to increase pressure on
land use for supply of other crops. This implies that sugarcane supply increase is majorly due
to increased sugarcane yields. Sugarcane yields increase by 11.06% between 2011 and 2030.
Without any major expansion in area under sugarcane, only 28.4 MT of sugar is produced
which is incapable of meeting the projected sugar demand of about 33 MT in 2030 leading to
price increase of sugar. Price of molasses is expected to increase by almost 54% to Rs. 9608
per tonne in comparison to 2011 levels and sugarcane price is around Rs. 2847 per tonne. A
rise in molasses prices raises the molasses ethanol prices (at refinery gate) to Rs. 51.62 per
litre.
5.3.2 Molasses and sugarcane scenario: Under the combined molasses and sugarcane
ethanol case, no significant increase in area under sugarcane is observed in comparison to
baseline scenario. In absence of any significant expansion of area under sugarcane, diverting
only 10% of sugarcane by the major sugarcane producing states leaves only to 255.3 MT of
sugarcane for sugar production (which is almost 63% of total sugarcane produced), a fall of
8.3% in comparison to the baseline. As a consequence of reduced sugarcane supply for sugar
domestic sugar production falls by 9% leading to increase in prices of sugar by about 16.6% in
comparison to the baseline case. Accordingly, molasses production also decreases by about
33.7% to 8.5 MT which equally impacts molasses ethanol supply in comparison to the baseline
scenario. However, total ethanol supply in this scenario increases by about 28% to 3596.6
million litres in comparison to the baseline case. The ethanol mix is 1861.7 million litres of
molasses ethanol and 1734.9 million litres is sugarcane ethanol. The maximum achievable
blending rate through mix of molasses and sugarcane feedstock is 3.7% which is
comparatively higher than the baseline case. Increased blending rate however comes at the
cost of loss in domestic sugar supply and higher prices of sugar, which may raise food security
concerns. The optimal feedstock mix comprises of 23 million tonnes of sugarcane (about 6%
of total sugarcane produced) and about 8.5 million tonnes of molasses. Even with a
combination of molasses and sugarcane ethanol meeting 20% blending mandate by 2030 is
challenging unless there is large area expansion under sugarcane. Feedstock prices i.e.,
molasses and sugarcane prices under this scenario are Rs. 9609 per tonne and Rs. 3133 per
tonne respectively. Increase in sugarcane requirement for both ethanol and sugar has positive
impact on prices since sugarcane price is almost 10% higher than the baseline scenario. The
refinery gate price of sugarcane ethanol is Rs. 48.3 per litre and molasses ethanol is Rs. 51.6
per litre.
Analysis of both the scenarios with first generation feedstock suggest that it is impossible to
meet increasing biofuels demand from only molasses feedstock and even with a mix of
sugarcane and molasses feedstock in 2030, without any significant expansion in sugarcane
acreage. In a previous study, Khanna et al., (2013) found that achieving 20% blending mandate
by 2017 with a mix of molasses and sugarcane feedstock is possible without any significant
food and land use impacts. However, our study results indicate that gasoline consumption
growth rate is expected to surpass ethanol consumption growth rate as a result of which
despite of increasing ethanol consumption in comparison to 2011 level, the ethanol blending
rate will be decreasing by 2030. It will be difficult to sustain 20% blend rate or even 10% blend
rate without any diversification of feedstock sources. Therefore, for India to achieve the
mandate by 2030 it is necessary to complement the existing ethanol feedstock basket with
second generation feedstocks.
5.3.4 All feedstocks scenario: Under this scenario, when all feedstocks (molasses, sugarcane
juice and cellulosic biomass is considered for ethanol production, the ethanol supply increases
substantially to 19301 million litres which helps achieve 20% blending mandate by 2030. The
biofuel mix under this case comprises of 2580 million litres of molasses ethanol, 1735 million
litres of sugarcane ethanol, and 14985 million litres of cellulosic ethanol. In total cellulosic
ethanol, bagasse and wheat straw ethanol are the major contributors 13442 million litres and
1532 million litres respectively and almost nil ethanol from rice straw. This is most likely due
to relatively lower ethanol recovery rate (by assumption ethanol recovery rate from rice straw
is almost 15% lower than bagasse and 4% lower than wheat straw) and higher processing cost
of ethanol from rice straw in comparison to other cellulosic feedstocks (by assumption
ethanol processing cost from rice straw is almost 9.2% higher than bagasse ethanol and about
6% higher than processing cost of wheat straw ethanol). Producing 14974 million litres of
cellulosic ethanol, approximately requires 51.3 MT of biomass. Optimal feedstock mix under
this scenario is 45.6 million tonnes bagasse, 5.7 million tonnes of wheat straw, 11.7 million
tonnes of molasses and about 23 million tonnes of sugarcane. Wheat straw harvested area is
about 23% of wheat acres while bagasse requirement is met from total area under sugarcane.
Biomass price under this scenario turns out to be Rs. 2,548 per tonne. Comparing with the
baseline case, the results suggest that sugarcane area expansion falls by 0.3% to 5.12 Mha. In
fact, area expansion under sugarcane is negligible. However, due to low flexibility in cropland
expansion the domestic sugar supply is insufficient to meet the domestic sugar requirement
as a result of which sugar prices are 14% higher in comparison to the baseline case. Additional
ethanol supply from cellulosic feedstocks in 2030 eases demand pressure on molasses for
biofuel and industrial use purposes leading to fall in molasses prices (4%) in comparison to
baseline. Lower molasses prices reduce the refinery gate price of molasses ethanol by about
3.4% to Rs. 49.4 per litre. In similar condition, refinery gate price of sugarcane ethanol is Rs.
49 per litre and cellulosic ethanol is Rs. 48.3. Further, in the case of ‘All feedstock’ since
ethanol displaces considerable amount of petrol, the price of petrol marginally falls by 0.7%
due to reduced demand for petrol.
 Fig. 4: Scenario-wise Optimal Feedstock Mix 2030
60 54
50

40

30 24 24
20 14 12 11
10

0
Only Molasses Combined Molasses & All-Feedstock
Sugarcane

Molasses Sugarcane Biomass

Source: Author’s Estimates

Fig. 5: Scenario-wise Optimal Ethanol Mix 2030

18000
16000
14000
12000
10000
8000
6000
4000
2000
0
Only Molasses Combined Molasses & All-Feedstock
Sugarcane

Molasses Ethanol Sugarcane Ethanol Cellulosic Ethanol

Source: Author’s estimates

 Figure 6: Scenario-wise feedstock prices 2030
12000

10000 9609 9609

9129

8000

6000

4000 3133 3182

2575
2000

0
Only Molasses Combined Molasses & All-Feedstock
Sugarcane

Molasses Sugarcane Biomass

Source: Author’s estimates

5.4. Welfare impacts of alternate feedstock mix

As compared to the baseline case, the ‘Combined Molasses and Sugarcane Ethanol’ leads to
marginal increase in income gain (0.02%) for fuel producers and no change in income for fuel
consumers. Under the ‘All-feedstock’ scenario income gains for fuel producers marginally
increases (0.1%). As far as fuel consumers are concerned, fuel consumers is expected
experience economic gains (0.5%) mainly due to comparatively lower blended fuel prices.
Agricultural producers are expected to attain significant gains under both ‘Combined
Molasses and Sugarcane’ scenario (28%) and ‘All-Feedstock’ scenario (28%) in comparison to
baseline case. This is mostly due to rise in prices of crops like sugarcane, sugar and additional
income from biomass sale and the spill over effects of these on other commodity prices. In
the ‘All feedstock scenario’, cellulosic biomass producers’ potential gain from sale of residues
is about Rs. 24 billion6. This indicates that use of cellulosic biomass for ethanol can generate
significant economic gains and can also help achieve the 20% blending mandate by 2030.
Agricultural consumer surplus is the lowest in the ‘All Feedstock scenario’ due to
comparatively higher prices of agricultural commodities like rice and groundnut in
comparison to the baseline case.

In summary, total economic surplus would increase the maximum in the ‘Combined Molasses
& Sugarcane’ relative to the other two scenarios. The key difference under different scenarios
is the income distribution between different producer and consumer groups. The main driver

6
Cellulosic biomass includes all three biomass sources i.e. rice straw, wheat straw and bagasse. Economic gain
of cellulosic biomass producers is represented by difference between revenue earned from residues and residue
costs determined at the equilibrium prices and quantities.
for changes in economic welfare is the changes in agriculture producer and consumer surplus
and government revenue earnings. Under the ‘All feedstock scenario’, government revenues
decline significantly (14%) due to fall in tax revenues from reduced consumption of petrol in
total fuel demand (as 20% of petrol consumption is replaced with ethanol in total fuel
consumption). State and central taxes on petrol together accounted for about 39% of the
retail selling prices f petrol in the base year.

6. Conclusion
In this study we present an analysis of projected agriculture and fuel market conditions in
India and compared a business-as-usual regime with two alternate scenarios engaging
different combinations of feedstocks while maintaining the blending mandates. We use a
price endogenous mathematical programming model to analyse the impacts of alternate
feedstock combinations on food, land use and economic surpluses of food and fuel consumers
and producers.
The main findings of our study suggest impossibility of meeting increasing biofuels demand
from only molasses feedstock and even with a mix of sugarcane and molasses feedstock in
2030, without any significant expansion in sugarcane acreage. The scope of area expansion
for sugarcane is highly restricted due to high demand for other crops like rice, wheat, and
maize. Since gasoline consumption growth rate is expected to exceed ethanol consumption
growth rate the study finds that although ethanol consumption is expected to increase in
comparison to 2011 level, the ethanol blending rate will be decreasing by 2030. It will be
difficult to comply 20% blend rate or even 10% blending requirement by 2030 without any
significant diversification of feedstock sources.
We find that a feedstock combination of molasses, sugarcane and cellulosic biomass can help
India achieve 20% mandate by 2030. With limited scope for area expansion, cellulosic ethanol
can potentially help dampen the pressure of land expansion for sugarcane crop as compared
to situations where the mandate is met with only sugarcane-based feedstocks. Our results
indicate that cellulosic biomass producers (farmers) can enjoy significant economic benefits
from sale of agricultural residues although replacing petrol with ethanol can lead to significant
reduction in government revenues since, tax incidence on petrol in India is very high.
Cellulosic ethanol can potentially contribute 78% of total ethanol mix with bagasse as the
main residue source. Complementing the ethanol mix with cellulosic feedstock also prevents
prices surge of other ethanol feedstocks i.e., molasses and sugarcane.
Biofuel impact assessment literature is well documented with biofuel induced food security
and land use impacts dampening the biofuel prospects in developing countries. The results of
our study have important policy implications. Our study highlights that for countries like India,
where land under agriculture is nearly stagnating and forest land diversion for biofuels is
regulated by law, there will possibly be only two cases i.e., missing the mandates or achieving
the mandate through different feedstock combinations and ethanol conversion technologies.
Our study does not suggest unrealistic diversion of pastureland and forestland for biofuel
production expansion. Therefore, this study provides useful insights in developing country
context. However, an important point to note here is that the model simulation procedure is
an iterative process and the findings will get finer with improved data. Therefore, the results
of our study should not be interpreted as forecasts but as potential biofuel development
pathways for India under the assumptions considered for respective scenarios. The model
developed and used in this study can be expanded with several extensions which is of
importance for further in-depth assessment of biofuel policies for India. The model used for
simulation is deterministic as it fails to incorporate changes in key climate variables (i.e.,
temperature and rainfall) in the agricultural sector model. The model can be enhanced into a
stochastic model by incorporating rainfall and temperature components to reflect the
variability in crop yields. Extensions to the model can also be made by including other
agriculture biomass sources like corn stover, cotton stalks etc.
Further, for cellulosic feedstocks we assumed an efficient supply chain for biomass collection,
transportation, storage and use for ethanol conversions. In addition to these factors, biomass
utilization for fuel in India will necessarily depend upon alternate residue uses. Currently,
there are lots of uncertainty in accurate estimation of alternate biomass requirements and
surplus agricultural residue availability for fuel conversions. In this direction, detailed spatial
assessment of biomass supply potential can be a significant research contribution. Further,
advanced biofuel conversion technologies in India are still in the nascent stage of
development and far from commercial reality. Under such conditions, uncertainties in
conversion costs and ethanol recovery rates is quite prominent. Although the study results
indicate high potential of cellulosic biomass in augmenting India’s biofuel capacity,
advancement in cellulosic biofuels in India, to a large extent will depend upon the
predictability of the policies. While the current policy environment for advance biofuels in
India is positive, immature residue and biofuel market conditions; high ethanol production
costs will be the major influencing factors in achieving the future mandates.

References

MoPNG, 2018. Indian Petroleum and Natural Gas Statistics, 2017-18.

Msangi, S., & Evans, M. (2013). Biofuels and developing economies: is the timing right?.
Agricultural Economics, 44(4-5), 501-510.
Findlater, K. M., & Kandlikar, M. (2011). Land use and second-generation biofuel feedstocks:
the unconsidered impacts of Jatropha biodiesel in Rajasthan, India. Energy Policy, 39(6),
34043413.
Ray, S., Goldar, A., & Miglani, S. (2012). The ethanol blending policy in India. Economic and
Political Weekly, 23-25.
Saravanan, A. P., Mathimani, T., Deviram, G., Rajendran, K., & Pugazhendhi, A. (2018). Biofuel
policy in India: a review of policy barriers in sustainable marketing of biofuel. Journal of
cleaner production, 193, 734-747.
Biofuel Policy 2009, Ministry of Peroleum and Natural Gas.
Ray, S., Miglani, S., Goldar, A., 2011a. Ethanol Blending Policy in India: Demand and Supply
Issues. ICRIER policy series.
Kumar, S., Chaube, A., & Jain, S. K. (2012). Critical review of jatropha biodiesel promotion
policies in India. Energy Policy, 41, 775-781.
Axelsson, L., Franzén, M., Ostwald, M., Berndes, G., Lakshmi, G., & Ravindranath, N. H. (2012).
Jatropha cultivation in southern India: assessing farmers' experiences. Biofuels, Bioproducts
and Biorefining, 6(3), 246-256.
Ariza-Montobbio, P., & Lele, S. (2010). Jatropha plantations for biodiesel in Tamil Nadu, India:
Viability, livelihood trade-offs, and latent conflict. Ecological Economics, 70(2), 189195.
Aradhey, A. (2019). India: Biofuels Annual, GAIN Report No. IN9069, USDA Foreign Agricultural
Service.
Lavanya, B. T., & Manjunatha, A. V. (2018). The grey shades of sugar policies in India. Economic
and Political Weekly, 53(46), 36-44.
Hiloidhari, M., Das, D., & Baruah, D. C. (2014). Bioenergy potential from crop residue biomass
in India. Renewable and sustainable energy reviews, 32, 504-512.
Cardoen, D., Joshi, P., Diels, L., Sarma, P. M., & Pant, D. (2015a). Agriculture biomass in India:
Part 1. Estimation and characterization. Resources, Conservation and Recycling, 102, 39-48.
Woltjer, G., & Smeets, E. (2016). 11 Biofuel Commitments in India and International Trade.
International Trade and Food Security: The Future of Indian Agriculture, 150.
Gunatilake, H., Roland-Holst, D., & Sugiyarto, G. (2014). Energy security for India: Biofuels,
energy efficiency and food productivity. Energy Policy, 65, 761-767.
Khanna, M., Önal, H., Crago, C. L., & Mino, K. (2013). Can India meet biofuel policy targets?
Implications for food and fuel prices. American Journal of Agricultural Economics, 95(2),
296302.
Samuelson, P. A. (1952). Spatial price equilibrium and linear programming. The American
economic review, 42(3), 283-303.
McCarl, B. A., & Spreen, T. H. (1980). Price endogenous mathematical programming as a tool
for sector analysis. American Journal of Agricultural Economics, 62(1), 87-102.
Takayama, T., & Judge, G. G. (1971). Spatial and Temporal Price and Equilibrium Models.
Amsterdam, North.
Takayama, T., & Judge, G. G. (1964a). Equilibrium among spatially separated markets: A
reformulation. Econometrica: Journal of the Econometric Society, 510-524.
Takayama, T., & Judge, G. G. (1964b). An intertemporal price equilibrium model. Journal of
Farm Economics, 46(2), 477-484.
Takayama, T., & Judge, G. G. (1964c). Spatial equilibrium and quadratic programming.
American Journal of Agricultural Economics, 46(1), 67-93.
Chen, X., Huang, H., Khanna, M., & Önal, H. (2014). Alternative transportation fuel standards:
Welfare effects and climate benefits. Journal of Environmental Economics and Management,
67(3), 241-257.
Nuñez, H. M., Önal, H., & Khanna, M. (2013). Land use and economic effects of alternative
biofuel policies in Brazil and the United States. Agricultural Economics, 44(4-5), 487-499.
McCarl, B. A. (1982). Cropping activities in agricultural sector models: a methodological
proposal. American Journal of Agricultural Economics, 64(4), 768-772.
Önal, H., & McCarl, B. A. (1989). Aggregation of heterogeneous firms in mathematical
programming models. European Review of Agricultural Economics, 16(4), 499-513.
Bhan, S., & Behera, U. K. (2014). Conservation agriculture in India–Problems, prospects and
policy issues. International Soil and Water Conservation Research, 2(4), 1-12.

Cardoen, D., Joshi, P., Diels, L., Sarma, P. M., & Pant, D. (2015b). Agriculture biomass in India:
Part 2. Post-harvest losses, cost and environmental impacts. Resources, Conservation and
Recycling, 101, 143-153.
Tripathi, A. K., Iyer, P. V. R., Kandpal, T. C., & Singh, K. K. (1998). Assessment of availability and
costs of some agricultural residues used as feedstocks for biomass gasification and briquetting
in India. Energy conversion and management, 39(15), 1611-1618.
De Gorter, H., Drabik, D., & Just, D. R. (2015). The economics of biofuel policies: impacts on
price volatility in grain and oilseed markets. Springer.
Dahl, C. A. (2012). Measuring global gasoline and diesel price and income elasticities. Energy
Policy, 41, 2-13.
Shinoj, P., Raju, S. S., Chand, R., Kumar, P., & Msangi, S. (2011). Biofuels in India: future
challenges. Policy Brief, 36.
Raju, S. S., Parappurathu, S., Chand, R., Joshi, P. K., Kumar, P., & Msangi, S. (2012). Biofuels in
India: Potential, Policy and Emerging Paradigms. Policy Paper No. 27. National Centre for
Agricultural Economics and Policy Research. New Delhi-110 012.

Ray, S., Miglani, S., Goldar, A., 2011b. Ethanol Blending Policy: Issues Related to Pricing. ICRIER
policy series.
Purohit, P., & Fischer, G. (2014). Promoting Low Carbon Transport in India-SecondGeneration
Biofuel Potential in India: Sustainability and Cost Considerations.
Indian Sugar Mills Association
Das, S. (2020). The National Policy of biofuels of India–A perspective. Energy Policy, 143,
111595.
Price Policy for Sugarcane, 2017-18, Sugar Season, Commission for Agricultural Costs and
Prices, Ministry of Agriculture and Farmers, Welfare, Government of India.
National Policy on Biofuels, 2018. Ministry of Petroleum and Natural Gas, Government of
India.
Fajardo, D., McCarl, B. A., & Thompson, R. L. (1981). A multicommodity analysis of trade policy
effects: the case of Nicaraguan agriculture. American Journal of Agricultural Economics, 63(1),
23-31.
Dahl, C. A. (2012). Measuring global gasoline and diesel price and income elasticities. Energy
Policy, 41, 2-13.

Appendix
The Mathematical Model
Maximize
𝐷𝐸𝑀𝑐𝑜𝑚 𝐷𝐸𝑀𝑠𝑢𝑔 𝐷𝐸𝑀𝑎𝑙𝑐 𝐸𝑋𝑃𝑐𝑜𝑚
∑𝑐𝑜𝑚 ∫0 𝑓 (. )𝑑(. ) + ∫0 𝑓(. )𝑑(. ) +∫0 𝑓(. )𝑑(. ) + ∑𝑐𝑜𝑚 ∫0 𝑔(. ) 𝑑(. ) +
𝐸𝑋𝑃 𝐸𝑋𝑃 𝐼𝑀𝑃 𝐼𝑀𝑃
∫0 𝑠𝑢𝑔 𝑔(. )𝑑(. ) + ∫0 𝑎𝑙𝑐 𝑔(. )𝑑(. ) − ∑𝑐𝑜𝑚 ∫0 𝑐𝑜𝑚 𝑔 (. )𝑑(. ) − ∫0 𝑠𝑢𝑔 𝑔(. )𝑑(. ) −
𝐼𝑀𝑃
∫0 𝑎𝑙𝑐 𝑔(. )𝑑(. ) − ∑𝑎,𝑟 𝑐𝑜𝑐𝑎,𝑟 𝐿𝐴𝑁𝐷𝑟,𝑎 − ∑𝑟,𝑐𝑟 𝑟𝑠𝑟,𝑐𝑟 𝐿𝐴𝑁𝐷𝑅𝐸𝑆𝑟,𝑐𝑟

− 𝑐𝑓𝑠𝑠𝑐 𝜃𝑠𝑠𝑐 𝐶𝐴𝑁𝐸𝑠𝑢𝑔 − 𝑐𝑓𝑎𝑚𝑜𝑙 𝜃𝑎𝑚𝑜𝑙 𝑀𝑂𝐿𝐴𝑆𝑆𝑎 − 𝑐𝑓𝑒𝑚𝑜𝑙 𝜃𝑒𝑚𝑜𝑙 𝑀𝑂𝐿𝐴𝑆𝑆𝑒 − 𝑐𝑓𝑒𝑠𝑐 𝜃𝑒𝑠𝑐 𝐶𝐴𝑁𝐸𝑒
− 𝑐𝑓𝑒𝑟𝑒𝑠 𝜃𝑒𝑟𝑒𝑠 𝑅𝐸𝑆𝐼𝐷𝑈𝐸𝑒
-------------------------- (1)
𝐷𝐸𝑀𝑐𝑜𝑚 + 𝐸𝑋𝑃𝑐𝑜𝑚 ≤ ∑𝑎 𝑦𝑐,𝑎,𝑟 𝐿𝐴𝑁𝐷𝑎,𝑟 + 𝐼𝑀𝑃𝑐𝑜𝑚 ∀ 𝑐 ---------------- (2)

𝐷𝐸𝑀𝑠𝑢𝑔 + 𝐸𝑋𝑃𝑠𝑢𝑔 ≤ 𝑐𝑓𝑠𝑠𝑐 𝑆𝐶𝐴𝑁𝐸𝑠 + 𝐼𝑀𝑃𝑠𝑢𝑔 𝑓𝑜𝑟 𝑠𝑢𝑔 ---------------------- (3)

𝑆𝐶𝐴𝑁𝐸𝑆 + 𝑆𝐶𝐴𝑁𝐸𝐸 ≤ 𝑦𝑠𝑐,𝑟 𝐿𝐴𝑁𝐷𝑠𝑐,𝑟 -------------------------- (4)

𝐷𝐸𝑀𝐴𝑙𝑐 + 𝐸𝑋𝑃𝐴𝑙𝑐 ≤ 𝑐𝑓𝑎𝑚𝑜𝑙 𝑀𝑂𝐿𝐴𝑆𝑆𝑎𝑙𝑐 + 𝐼𝑀𝑃𝑎𝑙𝑐𝑜 (Alcohol balance) ---------------- (5)

𝑠𝑐
𝑀𝑂𝐿𝐴𝑆𝑆𝑒 + 𝑀𝑂𝐿𝐴𝑆𝑆𝑎𝑙𝑐 = 𝑐𝑓𝑚𝑜𝑙 𝑆𝐶𝐴𝑁𝐸𝑠
------------------------(6)
𝐸𝑇𝐻𝐷 = 𝑏𝑙𝑒𝑛𝑑. 𝐹 --------------------- (7)

𝐸𝑇𝐻𝐷 = 𝐸𝑇𝐻𝑐𝑎𝑛𝑒 + 𝐸𝑇𝐻𝑚𝑜𝑙𝑎𝑠𝑠 + 𝐸𝑇𝐻𝑏𝑖𝑜𝑚𝑎𝑠𝑠 -------------------------- (8)

𝐸𝑇𝐻𝑚𝑜𝑙𝑎𝑠𝑠 ≤ 𝑐𝑓𝑒𝑚𝑜𝑙 𝑀𝑂𝐿𝐴𝑆𝑆𝑒 ----------------------------- (9)

𝐸𝑇𝐻𝑐𝑎𝑛𝑒 ≤ 𝑐𝑓𝑒𝑐𝑎𝑛𝑒 𝑆𝐶𝐴𝑁𝐸𝑒 -------------------------------- (10)

𝐸𝑇𝐻𝑏𝑖𝑜𝑚𝑎𝑠𝑠 ≤ 𝑐𝑓𝑒𝑏𝑖𝑜𝑚𝑎𝑠𝑠 𝑅𝐸𝑆𝐼𝐷𝑈𝐸𝑒 ------------------------- (11)

∑𝑎 𝛿𝑎,𝑡,𝑟 𝐿𝐴𝑁𝐷𝑎,𝑟 ≤ 𝑙𝑟 ∀ 𝑟, 𝑡 --------------------------------------- (12)

𝑅𝐸𝑆𝐼𝐷𝑈𝐸𝑒 ≤ ∑𝑟,𝑐𝑟 𝑦𝑟𝑟,𝑐𝑟 𝐿𝐴𝑁𝐷𝑅𝐸𝑆𝑟,𝑐𝑟 -------------- (13)

𝐿𝐴𝑁𝐷𝑅𝐸𝑆𝑟,𝑐𝑟 ≤ 𝐿𝐴𝑁𝐷𝑎,𝑟 ----------------------- (14)

∑𝑎 𝐿𝐴𝑁𝐷𝑎,𝑟 = ∑𝑦 𝜆𝑦,𝑟 𝐻𝑖𝑠𝑡𝐴𝑟𝑒𝑎𝑦,𝑟,𝑎 ∀ 𝑟, 𝑐 ---------------------- (15)

∑𝑦 𝜆𝑦,𝑟 ≤ 1 ∀ 𝑟------------------------------ (16)

𝐿𝐴𝑁𝐷𝑎,𝑟 , 𝐿𝐴𝑁𝐷𝑅𝐸𝑆𝑟,𝑐𝑟 , 𝐷𝐸𝑀𝑐𝑜𝑚 , 𝐸𝑋𝑃𝑐𝑜𝑚 , 𝐼𝑀𝑃𝑐𝑜𝑚 , 𝐷𝐸𝑀𝑠𝑢𝑔 , 𝐸𝑋𝑃𝑠𝑢𝑔 , 𝐼𝑀𝑃𝑠𝑢𝑔 , 𝑅𝐸𝑆𝐼𝐷𝑈𝐸𝑒 ,
𝑆𝐶𝐴𝑁𝐸𝑆 , 𝑆𝐶𝐴𝑁𝐸𝐸 , 𝐷𝐸𝑀𝐴𝑙𝑐𝑜 , 𝐼𝑀𝑃𝑎𝑙𝑐𝑜 , 𝑀𝑂𝐿𝐴𝑆𝑆𝑎𝑙𝑐𝑜 , 𝑀𝑂𝐿𝐴𝑆𝑆𝑒 , 𝐸𝐷, 𝐸𝑇𝑐𝑎𝑛𝑒 , 𝐸𝑇𝑚𝑜𝑙𝑎𝑠𝑠 , 𝐸𝑇𝑏𝑖𝑜𝑚𝑎𝑠𝑠 ,
𝑀𝑂𝐿𝐴𝑆𝑆𝑒 ≥ 0 --------------------- (17)

where, 𝑓(. ) represents the inverse demand function, 𝑑(. ) denotes the integration operator and 𝑔(. )
is represents the export-import functions for the traded goods. The objective function (1) represents
the area under the demand functions and area under the gasoline supply function up to the quantities
in the optimum solution minus the cost production of crops, and the processing cost of sugar and
ethanol. Equation (2) is the material balance constraint for crop commodity stating that total demand
for crop commodities including exports to the rest of the world (ROW) should be less than or equal to
total domestic supply and imports from the ROW. Similarly, Equation (3) specifies the material balance
constraint for sugar. Equation (4) is the material balance constraint for sugarcane, indicating that the
sum of sugarcane used for sugar production and ethanol production should be less than total supply
of sugarcane. Equation (5) is the material balance constraint for processed commodity i.e., alcohol
indicating that the demand for alcohol is restricted to the supply of alcohol from molasses, which is
the sum of domestic alcohol production and imported alcohol from ROW. Equation (6) states that the
total molasses allocated to ethanol and alcohol production should be equal to the domestic supply of
molasses generated from cane processing for sugar. Equation (7) states the policy constraint for
ethanol blending. Equation (8) represents the ethanol demand-supply balance which states that the
total ethanol demand cannot exceed the total ethanol supply which is composed of three ethanol
types i.e., ethanol from sugarcane, ethanol from molasses and ethanol from biomass. Equation (9)
provides the molasses ethanol supply balance which states that the total quantity of molasses-based
ethanol produced should be restricted by total domestic molasses ethanol supplied from total
molasses used for ethanol production proportional to the conversion factor of molasses to ethanol.

Similarly, equation 10 specifies the sugarcane ethanol supply balance which states that the total
domestic supply of sugarcane-based ethanol should be restricted to total domestic supply of
sugarcane ethanol produced from sugarcane diverted to ethanol production proportional to the
conversion factor of sugarcane to ethanol. Equation 11 states that the total cellulosic ethanol
produced must be produced from total quantity of biomass available for ethanol production
proportional to the conversion factor of biomass to ethanol. Equation (12) is the land availability
constraint which specifies that the total land allocated to different crops in a region is limited by the
total available land in that region. Equation 13, states that the total supply of crop residues used for
cellulosic ethanol production must match the sum of regional supply of crop residues which is a
function of crop residue yield specific to crop in region and land available for residue collection in
regions specific to crop for residue. Equation 14 requires that the land from which crop residues are
collected cannot exceed the land allocated to crops specifically producing residues. Equation (15) and
(16) provides the production possibility frontiers for farmers’ acreage allocation decisions. Both the
constraints determine the planting flexibility constraint which is specified by historical mix constraint.
The R.H.S of the equation is the sum of historical acreages where is the weight variable for historical
crop-mix. The weight variables are region specific and are endogenously determined by the model.
The equation states that in each region the land allocated to all crop producing activities must be a
weighted average of the historically observed acreages of that crop. In equation (15), represents the
cropping possibilities for crop in region. This allows the acreage of the historical mixes to float up and
down but requires that crops fall in a combination of historic and hypothetic proportions. Equation
(16) states that sum of the weights should be less than equal to 1 (convexity constraint). Equation (17)
is the non-negativity constraint for the decision variables of the model.